Curious musings on how the housing market might change

An analyst for Citgroup, Matt King, created what he proclaimed "the most depressing slide I've ever created" and presented it a few days ago:  

Real estate prices are straight forward (and they're for single family houses).  The dependency ratio is the proportion of people that are working age to those that aren't or, in other words, the ratio of people that can reasonably be expected to be working to those that can't.  As the proportion of the population that isn't working grows relative the proportion that is, which is currently happening in much of the developed world because of the aging workforce (birth rates are dropping, so the dependency ratio is not due to more babies being born), real estate prices drop and are expected to continue dropping (for a bunch of reasons, one of which being that as retirees downsize their housing needs and there are not as many younger buyers remaining to buy the house they're moving from, causing housing prices to drop).

I get that it's a pretty simple slide that shows some trends, and I furthermore understand that correlation is not causation.  As a home owner, it kind of scares me because most of my wealth is wrapped up in my house.  But as a student of economics as it applies to the construction industry, I cannot help wondering what this slide means for the future of building homes.  If the value of single family housing drops precipitously over the next decade or two, does that mean that single family homebuilding will also decrease on a relative basis?  This is a big deal, as the largest segment of the construction industry is in the single family sector, and as goes single family housing, so goes the rest of the construction market.

There may actually be a huge opportunity here, though.  If you are a builder that specializes in multi-family housing, either apartments or high-rises, this could be great news.  The population of the countries in the figure above, including the United States, isn't decreasing, it's just getting older.  People will still need housing, just maybe not single family homes.  The single family housing market is starting to heat back up, but that's due in large part to low mortgage rates, low supply and mass home buying by private equity investors (who will likely rent the homes, at least in the near term).  Longer term, the cost of capital is going to go up (it can't get much lower and quantitative easing cannot go on forever).  The conventional wisdom regarding home ownership as a solid investment is being challenged, notably by Nobel Prize winning economist Robert Shiller (see video here). And many people that would traditionally be the age of purchasing their first home are holding off on doing so (they're too busy paying off student loans).  This doesn't inspire a great deal of long-term confidence in the single family housing market.

But going back to what originally made me curious about this.  If there is an opportunity for the developers and builders of multi-family housing projects, then that may spillover into other necessary construction projects.  Like single family housing, multi-family housing will spur the construction of schools, retail space, hospitals, etc.  But it may also spur big infrastructure investments, such as mass transportation projects.  I'm guessing the face of housing will change dramatically over the next 20 years to the detriment to single family housing builders that fail to evolve, but that will create opportunities for other builders that are ready can respond to the change in the market.